Prostate cancer is (CaP) the most common non-dermatological malignancy and results in approximately 38,000 deaths annually. The clinical behavior of CaP clinical varies widely and we lack sufficiently sensitive predictors of disease behavior to appropriately treat each individual diagnosed with this disease. Although emerging technologies that broadly monitor cellular biology offer great promise in improving our ability to understand and predict CAP behavior, there remain significant challenges in the clinical application of new technologies to the care of men with CaP. I have worked with Dr. Todd Golub at the Dana Farber Cancer Institute (DFCI) and Whitehead Institute at MIT to apply microarray analysis of cellular expression to important questions in CaP biology and clinical behavior. We have tested this technology on an in vitro, androgen-responsive model of CaP, demonstrated the feasibility of obtaining expression profiles from primary and metastatic CaP tumors, and now are prepared to bring expression analysis out of the lab and into the clinic. There are both technical and clinical aims of this proposal. First, we will incorporate the techniques of laser capture microdissection (LCM) and referenced PCR signal amplification (RPSA) into our established methods of expression analysis in order to decrease the amount of primary tissue required for analysis from CaP tumors and broaden the number of patients who may eventually benefit from this technology. Second, we will determine if expression profiles from CaP tumors correlate with clinical outcome following radical prostatectomy. CaP tumors will analyzed with our established techniques as well as after LCM and RPSA to determine the best methods to correlate expression profiles with disease outcome. Finally, expression analysis will be incorporated into clinical trials to determine its ability to correlate expression analysis with response to a specific systemic intervention (neoadjuvant chemotherapy) and to identify patients who will benefit from adjuvant hormonal therapy following radical prostatectomy. These aims would not be possible without the resources available at the DFCI and Whitehead Institute at MIT where an active group of clinicians, molecular biologists, and experts in bioinformatics is dedicated to applying expression analysis to clinical care. This research proposal, together with didactic training in bioinformatics and clinical study design, will develop the range of skills necessary to apply the best available technologies to the most appropriate clinical dilemmas in the management of men with CaP.